Mobilization has published a number of important articles over its 25 year lifespan, but one contribution that I have found particularly useful is the 2006 Hess and Martin article on when repressive events may backfire. The starting point of their analysis is the simple observation that repression of dissent in some cases leads to greater mobilization when repression generates public outrage against the repressor. Repression may in this sense “backfire” and undermine the state, thus marking a transformative event for social movements. Hess and Martin argue that in order for repression to backfire information about the event must be revealed or be accessible to the relevant audience, and the repressive events must be perceived as unjust, as for example when peaceful protest is met by disproportionate violent repression. Hess and Martin illustrate their argument by detailed analysis of three case where repression backfired, including how support for independence in India increase following efforts to repress the 1930 Salt March.
Although the ambiguous consequences of repression have been stressed by many – indeed the mixed effectiveness of repression is highlighted as a so-called punishment puzzle in Davenport’s (2007:8) review of research on repression – Hess and Martin provide an insightful and innovative analysis of how states may seek to prevent repression from backfiring. They suggest five different tactics, including 1) efforts to prevent coverage or information about events, 2) discrediting targets to make repression seem less offensive, 3) recasting the presentation of events (for example, preventive repressive acts as defensive measures), 4) providing official legitimating accounts through institutions or inquiries, or 5) intimidating or bribing witnesses to events. Their analysis anticipates the rationale behind many of the measures taken against current protest movements, as seen in the recent internet blackout in Iran to prevent information about anti-government protest or efforts to link protests with hooliganism and crime in Hong Kong.
The Hess and Martin has been widely influential in subsequent analyses of repression, and as of the time of writing their article has over 120 recorded citations in published work in the Web of Science. However, I would argue that many of the most interesting implications of their article have yet to receive as much attention as they deserves. In particular, if we turn to the implications for expectations, their core argument can also help us understand why states often do not repress as well as when mobilization is more likely to be effective.
It is often assumed that states will respond to all serious challenges to the status quo with repression, and Davenport (2007: 7) highlights the so-called “law of coercive response” as one of the core findings of research on repression. However, efforts to code repression in response to specific events such as the Social Conflict Analysis Database suggest that about 75% of all events see no repressive response, which seems to contradict the claims about a pattern with a regularity resembling a law.
One might dismiss the relevance of the high share of events not repressed by arguing that these unrepressed events cannot have been too threatening to state, but without clear ex ante specifications of what makes events fail to be threatening this line of reasoning could potentially become circular. A more interesting answer to this puzzle is that the potential benefits of repressing a challenge often will be outweighed by various potential costs of repression. In particular, states often find it better not to respond to smaller challenges through repression in order not to draw more attention to the phenomenon, and even if it is not technically difficult to repress a small event it may often be counterproductive to do so. Consistent with the analysis of Hess and Martin, denying attention to movements is in itself an important way to prevent awareness of incipient movements and curb further mobilization. However, this also has implications indicating possible cases where there is no repression than could backfire, as opposed to the observed cases of repression that may or may not backfire emphasize in their article.
If we add expectations onto their original analysis of Hess and Martin it also becomes possible to think of not just cases where resort to repression will fail to quell non-violent dissent, but also cases where nonviolence works precisely because forward looking agents realize that repression is likely to backfire and therefore backs done before going down that route. Although Honecker is reported to have favoured a Chinese style response to protest in East Germany, others in the Politburo probably correctly anticipated the limitations of such an approach and moved to remove him from office. In other cases, repression may fail to happen if security forces anticipate a public backlash against them, as evidenced by police officers avoiding to carry out orders or the army either removing leaders or calling for them to step down, as seen recently in Algeria and Sudan. In this case, the scale of the protest generated a larger likely backfire effect, as discussed by Hess and Martin, but the anticipation of this will itself likely prevent resort to repression in the first place.
Hess and Martin explicitly present their analysis as a vehicle to generate hypotheses and actionable knowledge for activists. Beyond the implication they discuss, the article can also help draw our attention to how the expectation of when repression may backfire can play an important role in determining state responses and successful mobilization. At the same time, this leads to the classical problem highlighted in studies of deterrence, where success is hard to measure since we cannot clearly attribute the absence of violence to a strategy. But it may be possible to derive empirical approaches to the observable implications and controlled experiments to see if behavior can be shifted through features influencing expectations. I hope Mobilization can help provide an outlet for research of this type, in addition to theory building exercises in the spirit of Hess and Martin.